Amazon Bedrock

1. What is Bedrock

   Managed access to foundation models through a unified API.

2. Models at a glance

   Model providers, modalities and capabilities.

3. Model availability

   Regional availability, access and cross-Region inference.

4. Building with Bedrock

   Inference endpoints, APIs, tool use and projects.

5. Making inference requests

   InvokeModel, Converse and streaming responses.

6. Converse API

   Unified multi-turn conversation interface across models.

7. Tool use

   Function calling with models.

8. Model customization

   Fine-tuning, continued pre-training and distillation.

9. Guardrails

   Content filters, denied topics and sensitive information policies.

10. Batch inference

    Processing large prompt sets asynchronously.

11. Provisioned Throughput

    Reserved model capacity for consistent throughput.

12. Observability

    Invocation logging and CloudWatch metrics.

Each section links to the official page. Titles and summaries by Extra Credit; the docs belong to their authors.